Modelling Land-Use and Land-Cover Changes: A Hybrid Approach to a Coastal Area

Modelling Land-Use and Land-Cover Changes: A Hybrid Approach to a Coastal Area

Raquel Faria de Deus (Interdisciplinary Centre of Social Sciences (CICS.NOVA), NOVA FCSH, Universidade NOVA de Lisboa, Portugal), José António Tenedório (Interdisciplinary Centre of Social Sciences (CICS.NOVA), NOVA FCSH, Universidade NOVA de Lisboa, Portugal) and Jorge Rocha (Institute of Geography and Spatial Planning, Universidade de Lisboa, Portugal)
DOI: 10.4018/978-1-7998-2249-3.ch003
OnDemand PDF Download:
Available
$37.50
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

In this chapter, a hybrid approach integrating cellular automata (CA), fuzzy logic, logistic regression, and Markov chains for modelling and prediction of land-use and land-cover (LULC) change at the local scale, using geographic information with fine spatial resolution is presented. A spatial logistic regression model was applied to determine the transition rules that were used by a conventional CA model. The overall dimension of LULC change was estimated using a Markov chain model. The proposed CA-based model (termed CAMLucc) in combination with physical variables and land-use planning data was applied to simulate LULC change in Portimão, Portugal between 1947 and 2010 and to predict its future spatial patterns for 2020 and 2025. The main results of this research show that Portimão has been facing massive growth in artificial surfaces, particularly near the main urban settlements and along the coastal area, and reveal an early and intensive urban sprawl over time.
Chapter Preview
Top

Introduction

Sustainable land-use and land-cover (LULC) development requires increasingly more reliable and high-quality spatial information to guide smart spatial planning and land-use decision-making (Magarotto, Faria-De-Deus, Ferreira Costa, & Masanet, 2017; Rocha, 2019). This applies in particular to the geographical context of coastal areas, which have been undergoing rapid and strong urbanization dynamics in the past few decades. The fast population and economic growth along coastal areas presents both an opportunity and a challenge for the sustainability of coastal ecosystems and built environments (Freire, Santos, & Tenedório, 2009; Martinez et al., 2017). Presently, 2.5 billion people, i.e., 40% of the world’s population, live within 100 km of the coast (Burke, Reytar, Spalding, & Perry, 2011) adding great pressure to these productive and valuable coastal ecosystems. As coastal populations increase and natural coastal protection is reduced or lost, the sea level rise and local impacts of land-based stress factors, namely LULC change, occur together with global and regional stressors, such as climate change (Burke et al., 2011), resulting in altered dynamics and the subsequent reduced resilience of territories (Rocha, Ferreira, Simões, & Tenedório, 2007; Rocha, Gutierres, Gomes, & Teodoro, 2018).

In Portugal, as in the rest of the world, this has posed the challenge of rethinking the types of innovative actions that may contribute to more efficient LULC planning policies. Moving in this direction implies the application of state-of-the-art technologies to the studies targeting the analysis, monitoring, assessing and modelling of the process of LULC change in coastal areas (Boavida-Portugal, Rocha, & Cardoso Ferreira, 2016). The scientific advances that have been made in the field of urban growth and LULC change Cellular Automata (CA)-based simulation since the late 1990s have confirmed CA-based simulation as one of the most remarkable and promising geographic modelling approaches to interpret and represent the spatiotemporal complexity of LULC change, mainly due to their natural ability to capture and reproduce complex bottom-up processes through simple rules (Bone, Dragicevic, & White, 2011; Verburg et al., 2015; White, Engelen, & Uljee, 2015).

Research on LULC change CA-based modelling has since evolved a great deal, moving from its main focus on understanding the drivers and impacts of LULC change towards novel approaches (see, for example, Li & Gong, 2016; Gomes, Abrantes, Banos, & Rocha, 2019), which use this knowledge to properly develop sustainable land-use management practices and spatial policies. Even though CA-based models have already proven to be very effective at modelling and predicting large-scale LULC change, research on the development of fine-scale and long-term series LULC datasets to evaluate, quantify, and model LULC change, particularly at the local level, is still lacking. In order to produce more realistic urban phenomena simulations, CA-based models should improve their ability to simulate at a high level of spatial resolution. At the same time, increasingly complex urban spatial models that are very time-dependent require the use of physical and socioeconomic data from several time periods, since previous research has shown that the calibration process over time improves the results of simulation models (Liu, 2009). Thus, objectively approaching the analysis and modelling of LULC change at a fine spatial scale should be central for urban planning teams to efficiently define and conduct forthcoming land-use planning decisions.

Key Terms in this Chapter

Land-Use and Land-Cover Change: The physical expression of land surface change.

Coastal Area: The land area that extends landward of the coastline and that is directly or indirectly influenced by coastal processes. This is a more general term than coastal zone, which refers to the geographic area defined by the legislation as a zone for coastal management purposes.

Diffusion Cellular Automata Model: A model that reproduces the capacity of a cell to spread or propagate information over space through neighbouring cells.

Hybrid Modelling: A method that integrates the advantages of different techniques into one approach, while preserving accuracy.

Model: A simplified and approximate representation of a phenomenon, structure, or system of the real world.

Urban Phenomena Modelling: The approach concerned with developing and applying models in order to explore specific issues in urban studies by means of spatial computation techniques.

Spatial Determinant: A variable that may increase or detract from the suitability of a location for the presence of a particular land-use or land-cover category.

Transition Rule: Expresses the parameters and weights related with the spatial variables involved in a simulation and governs the result of its spatial evolution.

Land-Use and Land-Cover: The physical transposition of human activity on the earth’s surface, i.e., the territorial inscription of human activity.

Neighbourhood: In cellular automata, the neighbourhood of a cell is the set of nearby cells with which a specific cell interacts.

Complete Chapter List

Search this Book:
Reset